Medieval Britain and Ireland in 2011

Transcription

1 Medieval Archaeology, 56, 2012 Medieval Britain and Ireland in 2011 General Editor: NEIL CHRISTIE Portable Antiquities Scheme Report edited by: JOHN NAYLOR1 Medieval Britain and Ireland compiled and edited by: MÄRIT GAIMSTER and KIERAN O CONOR with RORY SHERLOCK This section of the journal comprises two core sets of reports linked to work in 2011: on finds and analyses relating to the Portable Antiquities Scheme (PAS) and on site-specific discoveries and reports in medieval Britain and Ireland (MB&I), with a selection of highlighted projects. For the PAS report, reviews on coin and non-coin finds and on specific research angles are presented. For MB&I, the Society is most grateful to all contributors (of field units, museums, universities, developers, specialist groups and individuals) who have provided reports on finds, excavations, field-surveys and building analyses for Note that, while we can advise on content, we are not able to abstract from interim reports. Please also note that in certain cases the National Grid Reference has been omitted from reports to protect sites; do notify the compilers if this information is to be withheld. For MB&I, see below for the format and content of the Fieldwork Highlights section and for the contact details of the compilers. The annual Specialist Groups Reports now appear in the Society s Newsletter. PORTABLE ANTIQUITIES SCHEME The Portable Antiquities Scheme (PAS) is a voluntary scheme to record archaeological finds made by members of the public. Many of the finds come from archaeological sites which have been damaged or eroded, usually by agriculture, leaving the objects as the only evidence of past activity; other objects are of interest in their own right. At the end of 2011, the PAS database contained 18,097 records of finds of early medieval date (c ad 410 c 1066) and 99,531 records of finds of high and later medieval date (c ad 1066 c 1500). A small proportion of these are subject to the Treasure Act 1996 which gives the Crown (in practice, museums) the right to acquire them; in 2011 there were 942 Treasure cases, of which 120 were of early medieval date and 224 were of high or later medieval date.2 In 2011, 82,615 finds (in 56,047 records) were made in total,3 of which 15,678 (14,225 records) were medieval. There were a total of 2,426 pre-conquest finds (in 2,042 1 Heberden Coin Room, Ashmolean Museum, Beaumont Street, Oxford OX1 2PH, England, UK; john.naylor@ ashmus.ox.ac.uk. 2 All finds reported via the Treasure Act 1996 are also now included on the Portable Antiquities Scheme Database (PASD). Full PAS and Treasure Annual Reports (PTAR) are freely available to download up to and including These can be downloaded from <finds.org.uk/treasure/reports>. 3 As of 31 December Note that a record on the database can include more than one find, especially for entries of Roman coinage and Treasure cases. 301 Society for Medieval Archaeology 2012 DOI: / Z

2 302 portable antiquities scheme records) and 13,252 post-conquest (in 12,183 records). Coinage accounted for 5,212 of these: 408 of pre-conquest date (282 single finds; 126 Treasure-related) and 4,804 post- Conquest (4,582 single finds; 222 Treasure-related).4 Full details of all finds recorded by PAS can be found on the Scheme s website. (J Naylor) focus on coinage in Early medieval The last year has again seen a large number of early medieval coins recorded, and PAS now holds a corpus of 2,326 non-hoard coins of this period, with 282 added in As in other years, besides forming significant new data, some specific finds are important additions to the corpus. The earliest finds of interest are the growing numbers of silver siliquae issued in the last decades of Roman Britain which were subsequently clipped (Fig 1a). This clipping probably started in the very late 4th century but appears to have continued well into the 5th century.7 Of the 192 siliquae recorded in 2011, 41 were listed as clipped and as such can be seen to illustrate the continued use of coin, in some manner, in the early post- Roman period. Furthermore, 15 more clipped coins and one gold solidus from a hoard were found near Mildenhall (Suffolk; SF-D767D4) in addition to two sets of previous finds.8 Alongside these, five 5th- to 7th-century Merovingian and Visigothic gold coins were recorded, including a pierced and gold-plated imitation of a Visigothic solidus dating to the late 5th century which copies a Ravenna solidus of Severus III (461 5) found at Southease (East Sussex; SUR-5B13A4; Fig 1b).9 Also recorded in 2011 were two early Anglo-Saxon gold coins (known as thrymsas) based on earlier Roman prototypes: one, a two emperors type, copying the design from 4th-century gold solidi, was found near Elmswell (Suffolk; SF-84A6C8; Fig 1c); the other from Wilton (Wiltshire; DEV-E8CCA1; Fig 1d) copies a coin of Crispus (317 26). Both coins represent part of the later phase of gold coinage from c Last year, PAS recorded more than 100 silver sceattas (c 670 c 760) in one year for the first time. These consolidate rather than extend the overall distribution of these coins, although a new find from Leominster (Herefordshire; HESH-B37EA8) of a Continental Series E porcupine type does add to the very small corpus known from W England.10 Two examples of very early sceattas can be noted, dating c , generally seen as transitional types between gold thrymsas and silver sceattas, of Pada type, from Dorking (Surrey; SUR-2CF753) and Barham (Suffolk; SF-0C2A17; Fig 1e). Twenty-seven middle and pre-reform late Anglo-Saxon pennies and 54 9th-century Northumbrian stycas were also recorded in 2011, the former almost double that from While all of these finds can be considered rarities, particularly interesting were the first find of a penny of Offa of Mercia (757 96) from Devon (Teignbridge; DEV-530DA3), a very rare issue of Beornwulf of Mercia (823 5) from Dorking (Surrey; SUR-8B14B0; Fig 1f) for which there are only c 22 other known examples, and one of few single finds of the moneyer Beagstan for Edward the Elder of Wessex ( ) found in Bythorn and Keyston (Cambridgeshire; NARC-3C43B1). Illustrating the pace at which the corpus of 4 A full breakdown of this numismatic data can be found in Allen et al forthcoming I must express my thanks and appreciation to all of the FLOs (Finds Liaison Officers) and other specialists whose hard work recording and identifying the large numbers of medieval coins is bringing such interesting and important material to light. 6 PASD, as of 31 December This total excludes any finds issued earlier but possibly re-used, such as the late-roman silver siliquae discussed below. 7 Abdy 2005, The two previous sets of coins recovered are: SF-9421A5 (2010 T73; 7 coins) and SF-4BDF20 (2007 T165; 24 coins). 9 The two pieces of this pendant were discovered five years apart, the first in 2006, the second in 2011 and no doubt originated as part of a burial assemblage. See also Naylor 2012 for an overview of gold coinage of this period. 10 Cf Naylor 2011a.

3 medieval britain and ireland fig 1 Early medieval coins. (a) IOW-F86317 clipped Roman siliqua. (b) SUR-5B13A4 imitation Visigothic solidus. (c) SF-84A6C8 Anglo-Saxon thrymsa. (d) DEV-E8CCA1 Anglo-Saxon thrymsa. (e) SF-0C2A17 Anglo-Saxon sceat. (f) SUR-8B14B0 penny of Beornwulf. (g) BH-7ACDB2 penny of Ecgberht. (h) NCL-544D22 Islamic dirham. (i) PUBLIC-8CC232 penny of Æthelred II. Scale 1:1. Photographs: (a) F Basford, (b) and (f) D Williams, (c) and (e) A Brown, (d) N Corey, (g) J Watters, (h) R Collins, (i) S Johnson. types for this period is growing is the fact that five of these coins were new types or variants of known types. Three of these are for Offa of Mercia (LIN , IOW-C8BD83 and NCL-AF9BE4) with one each for Ecgberht of Wessex (802 39; BH-7ACDB2; Fig 1g) and Æthelweard of East Anglia (c ; SF-793F43). Viking-period coinage finds were dominated by two major mixed hoards, the largest from near Silverdale (North Lancashire; LANCUM-65C1B4/2011 T569) and the other from the Furness area (Cumbria; LANCUM-80A304/2011 T283).11 Silverdale contained 174 silver objects and 27 silver coins within a lead container. It is typical of late 9th- to early 10th-century deposits, featuring a mix of Anglo-Saxon (including Viking imitations 11 The hoards are summarised in Boughton et al 2012.

4 304 portable antiquities scheme of Anglo-Saxon), Anglo-Scandinavian, Carolingian and Islamic coinage, with one coin recording what appears to be a previously unknown Viking ruler called Harthacnut, probably from around c 900. The Furness hoard was composed of 13 silver ingots plus 79 silver coins, with a deposition date in the 950s; the majority of the coinage of Anglo- Saxon origin was mostly in the name of Eadred (946 55). Alongside these major finds, a number of other finds of Anglo-Scandinavian pennies and Islamic dirhams were made, such as two complete dirhams from the Aldwincle area (Northamptonshire; DENO- 07CCA4) and Revesbury (Lincolnshire; NCL-544D22; Fig 1h); these are rarer in this country than the cut hacksilver fragments more typical of the dirham finds in Britain.12 Finally, 54 finds of post-reform Anglo-Saxon coins (ad ) were made in Of these, at least ten are cut fractions, mostly halfpennies (some finds were highly fragmented, thus preventing their full denomination to be determined), giving an indication of the development of coin use in this period. Prior to the late 10th century very few cut fractions or round halfpennies are known, but here we see the beginnings of a multi-denomination currency which has endured ever since. Most of the finds from 2011 were of standard types, but the coin of Cnut ( ) from Hayle (Cornwall; CORN- F52DB5) and two from Laugharne Township (Carmarthenshire; PUBLIC-F416D3 and PUBLIC-8CC232; Fig 1i) form evidence right at the edges of the known distributions of these coins. High and late medieval Post-Conquest coinage (ad ) again far outweighed pre-conquest coins in terms of quantity with 4,582 recorded in 2011 (an increase of over 10% on 2010), and continued to illustrate the extensive and varied circulation of coins in this period. The total of non-hoard post-conquest coins is now in excess of 30,000 records.13 For the earliest period (ad ), 97 coins were recorded, with nearly half being issues of Henry II s ( ) cross-and-crosslets type. Even within this small corpus there were a number of interesting new finds. A coin from Shepherdswell with Coldred (Kent; PUBLIC-C94F50) names a new moneyer for the mint at Hastings (East Sussex) under William I ( ) which has ramifications for our understanding of the levels of activity at the mint.14 Alongside this, 2011 saw good recovery of coins minted by Stephen ( ) and under the period of the Anarchy resulting in issues by competing groups: 15 coins of Stephen were recorded, mostly of the standard Watford type, but a rare type minted in Lincoln was found at Edingley (Nottinghamshire; LEIC-69AF01; Fig 2a);15 alongside these were two scarce issues of Henry, Earl of Northumberland ( ) found at Maulds Meabern (Cumbria; LANCUM-C7BD67) and Alnwick district (Northumberland; NCL-A6C172; Fig 2b).16 Foreign coins are also rare finds in this period, but two 11th-century Byzantine coins were recorded in 2011, one the first billon aspron trachy of Alexios I Komnenos ( ) recorded by PAS, from Milton Hill (Oxfordshire; FAJN-76BBF5; Fig 2c),17 the other an anonymous copper follis dating to the same reign from Beckenham (Greater London; SUR-8206C4) which fits the pattern of finds in London as witnessed from excavation finds.18 Other foreign coins included a denier of Bishop Conrad of Utrecht ( ), which had been modified for use as a 12 Naismith 2005, As at 31 December North 1994, no 845; two stars type (issued c ). 15 The Watford type (North 1994, no. 873) is the basis for a range of local variants, including this one from the eastern group and belongs to North 1994, no Both North 1994, no 913; Spink 2003, no 5012 (or variant thereof). NCL-A6C172 has now been purchased by the Society of Antiquaries of Newcastle upon Tyne, and is on display at the Great North Museum: Hancock in Newcastle upon Tyne. 17 Hendy 1999: DOC IV, no 25 (Second Coinage, post-1092 reform), Constantinople mint. 18 Class K anonymous follis, Second Coinage (pre-reform c ). For a brief overview of Byzantine coinage found in Britain, see Moorhead and Naylor For the finds from London, see Egan 2007.

5 medieval britain and ireland fig 2 High- and late-medieval coins. (a) LEIC-69AF01 penny of Stephen. (b) NCL-A6C172 penny of Henry, Earl of Northumberland. (c) FAJN-76BBF5 billon aspron trachy of Alexios I Komnenos. (d) NMS-66F9A5 brooch made utilising a denier from Utrecht. (e) DUR-88FA74 penny of Henry VI. (f) SUR-54DB38 half-noble of Philip the Bold, Duke of Burgundy. (g) BERK helmet denier of Bohemond III of Antioch. (h) SUSS-A584F5 soldino of the Republic of Genoa. Scale 1:1. Photographs: (a) W Scott, (b) R Collins, (c) J Naylor, (d) E Darch, (e) F McIntosh, (f) D Williams, (g) A Byard, (h) S Smith. brooch or badge, from Bradenham (Norfolk; NMS-66F9A5/2011 T103; Fig 2d). This resembles a type discussed here last year with the coin mounted to show the side with an image of a cathedral.19 A second modified coin, of William I, was found at Alverdiston (Wiltshire; SOM-F7FD62/2011 T 357).20 In the following period (ad ) the introduction of the short cross and voided long cross pennies coincided with a distinct increase in the circulation and distribution of coinage.21 This is reflected in the number of finds reported, with 1,370 new finds recorded in This expansion continued to the end of the medieval period 19 See Naylor 2011b (SWYOR-1DFAD6; 2010 T535). Although a single find, modified coins such as this constitute Treasure under the Treasure Act See <finds.org.uk/treasure> for further details. 20 Treasure case 2011 T357; see also Williams Allen et al 2012 forthcoming.

6 306 portable antiquities scheme (ad ) with 2,478 English and Irish coins recorded, and 118 Scottish coins.22 Notable were a rare Edward III ( ) farthing minted in Berwick-on-Tweed from Thwing (East Riding of Yorkshire; NCL-D6B492); an Anglo-Gallic demi-sterling (halfpenny) minted in Aquitaine (France) under Edward III from the Isle of Wight (IOW ), one of only six recorded by PAS;23 and, from Bedale (North Yorkshire; DUR- 88FA74; Fig 2e), a new penny for the mint at Calais under Henry VI ( ) which probably represents part of the last group of dies sent to Calais before it was lost to French control. Alongside these a range of archaeologically interesting finds was recorded in One of these was a halfgroat of Edward IV (first reign, ) found during excavations at the site of the Masonic Lodge, Northampton (Northamptonshire; NARC-03B111) as far back as The coin was found in the mouth of a skeleton and may have originally have been folded a practice which may relate to vows of pilgrimage.24 The placement of coins in burials in late-medieval England is a known, if rare practice, and may represent magical rites.25 Alongside this, more evidence for the modification of contemporary coinage into badges or brooches was seen, including a gilded and mounted gros tournois of Philip IV of France ( ) from North Duffield (North Yorkshire; YORYM-91F842) and a fragment of a gilded halfpenny of Edward III from Pleshy (Essex; ESS-9A1266), once mounted as a badge. The range of foreign coinage recorded in 2011 provides more evidence for international networks of trade. The bulk of finds reported were, as is usual, from Scotland, Venice and the Low Countries. One group of finds sheds light on international politics of the age: in 1469, England and Burgundy published a monetary agreement which made each other s currency legal tender in the other country;26 in this country, double patards of Charles the Bold ( ) are the most common finds, equivalent to the English groat, and 14 were recorded by PAS this year (taking the total to 93 on the PASD).27 Another Low Countries coin of importance recorded in 2011 was a gold half noble of Philip the Bold, Duke of Burgundy ( ) found at Hook (Hampshire; SUR-54DB38; Fig 2f), the first such coin recorded by PAS; this joins a very small corpus of late 14th-century foreign gold coins found in this country.28 Unusual finds from further afield included a coin of the Crusader states: a helmet denier of Bohemond III of Antioch ( ), found in Whatcombe (Berkshire, BERK ; Fig 2g), and a soldino from the Republic of Genoa (under Governor Valerad of Luxembourg) from 1397 was found at Plumpton (East Sussex; SUSS-A584F5; Fig 2h). Although Genoa was a major trading power in this period, finds of Genoese coinage in England are not common, being heavily outweighed by Venetian coins. (J Naylor) focus on non-numismatic finds in Early medieval Among the earliest Anglo-Saxon items recorded in 2011 was a hollow bone object (Fig 3a) with ring-and-dot decoration from Southwark (London; LON-E3F661). Close parallels come from late 4th- and 5th-century cremation burials in N Germany, where 22 English and Irish coins are counted together here as both were issued under the auspices of the English crown. 23 IOW is of Elias 1984 no 57, for which PAS has recorded one other (DOR-8D4A95). In addition, Cook 1999, 274 lists another five English finds of Anglo-Gallic issues. 24 See brief discussion of this in last year s round up (Naylor 2011b). Also see Kelleher Gilchrist 2008, Cook 1999, As of 31 December Cook 1999, 274 lists just nine examples of 14th-century foreign gold coin and none of Philip the Bold. This may be the first find of its type in this country. 29 My thanks must go to the FLOs and other specialists for their hard work in identifying and recording the large amounts of medieval objects brought into PAS every year. I must especially acknowledge the following for their help with identifications and parallels for objects included below: Ian Riddler, Catherine Hills, Barry Ager, Susan Youngs, Tim Pestell and Steven Ashley.

7 medieval britain and ireland 2011 they commonly contain sewing needles.30 Two undecorated examples occur at Spong Hill (Norfolk), but English finds are otherwise rare.31 Another 5th-century find is the terminal of a penannular brooch found near Loughborough (Nottinghamshire; LEIC-9CDC85). In the shape of an animal head, with circular eyes of inlaid silver, the top of the head shows a sub-circular panel with triskele decoration in relief. Most of the recesses retain decayed enamel. The animal holds a neatly finished hollow circular tube within its mouth, implying that the terminal was made separately from the rest of the brooch (Fig 3b). The closest parallel for the form of the terminal is a find now in North Lincolnshire Museum, said to be from Caistor (Lincolnshire; Fig 3c).32 Although made from silver, the Caistor fragment has a base-metal rod in its socket, showing how it was attached to the rest of the brooch in similar fashion to the Loughborough piece. Several Frankish imports were recorded in An incomplete 5th- or early 6thcentury copper-alloy bow brooch from the Isle of Wight (IOW-78CB34) is of a type known principally on the Continent, from northern France to the lower Rhineland. Belonging to Koch s type Bifrons/Preure Gruppe III.1.1.1,33 the animal heads at the upper corners of the foot are vestigial compared to those on other examples. Other English finds are known, including from Bifrons (Kent) and Great Chesterford (Essex) grave 9.34 Further continental objects include an early 6th-century copper-alloy scabbard-slide (Fig 3d) from Barham (Kent; KENT-B9EB67) typically found in the Rhineland and SW Germany;35 and a 6th-century copper-alloy brooch (Fig 3e) from Saleby (Lincolnshire; FAKL ) of Domburg type, usually found in Frisia, but with a close parallel from Wharram Percy (North Yorkshire).36 More mysterious is a gilded copper-alloy object (Fig 3g) from Somerton (Oxfordshire; BH-DB46C1). Barry Ager has suggested a possible identification as a hairpin of western Frankish origin, broadly paralleled by an early 6th-century gilded silver pin from Douvrend (Seine-Maritime) in the Musée des Antiquités at Rouen. This has a long head elaborately decorated with two chip-carved sea-creatures in procession, although their outlines form the edge of the head rather than being set inside a border.37 Metalworking dies feature more often in PAS records than in excavated assemblages, probably because they are rarely discarded and so are preserved only as accidental losses. Several were recorded in A matrix die from Boxford (W Berkshire; BERK- 8E4641) has a cross and dots engraved into the surface. It was probably used to produce a disc with imitation granulation and a linear design, such as the later 7th-century gold pendant from Wye Down (Kent).38 Another from Boxford (BERK-8CE912) is the more common patrix type 2, with the design in relief. Displaying three double-strand interlaced knots, it may date to the 7th or early 8th century. Of similar date is a third patrix die from Gatcombe (Isle of Wight; NARC-EE9CC3) which is in the shape of a predatory bird s head; the eye is made up of two further birds heads curving around to form a spiral. A fourth probable die was found near Alford (Lincolnshire; LIN-CD5B11). It has Eg Weber 2000, Spong Hill cremations 1976/2 (Hills and Penn 1981, 46 and fig 179) and 3078/2 (Hills et al 1994, 100, fig 128). 32 Accession number NOLMS: Koch 1998, 165 8, 697; Taf 26, 5 9, Karte The pair from Bifrons are not from Grave 34 (contra Koch 1998, 697) but are unstratified finds from unrecorded graves excavated by Lord Conyngham s gamekeeper in 1867; Hawkes 2000, 61, 70, fig 42, nos 68 and 69. For Great Chesterford, see Evison 1994, fig 18, pl 10e. 35 Menghin 1983, and B Ager in Milne and Richards 1992, 47 9, no. 14. This brooch has now been donated to The Collection, Lincoln. 37 Danet and Périn 1992, pl Webster and Backhouse (eds) 1991, no 37.

8 308 portable antiquities scheme fig 3 Early medieval objects. (a) LON-E3F661 needle-case. (b) LEIC-9CDC85 penannular brooch terminal. (c) Penannular brooch terminal from Caistor, Lincs. (d) KENT-B9EB67 scabbard-slide. (e) FAKL Domburg-type bow brooch. (f) BERK-9FA163 cross brooch. (g) BH-DB46C1 unidentified object. (h) LEIC-15A132 thimble-like object. Scale 1:1. Photographs and drawing: (a) K Sumnall, (b) W Scott, (c) and (e) K Leahy, (d) J Jackson, (f) A Byard, (g) J Watters, (h) P Rea. an animal in chip-carved relief within a tapering frame, with parallels on the nasal of the late 8th-century Coppergate (York) helmet.39 Other significant 8th-century finds in 2011 can be flagged. A gilded silver thimblelike object (Fig 3h) from Melton (Leicestershire; LEIC-15A132) bears animal art comparable to that on the linked pin set from the River Witham in Lincolnshire, sometimes 39 Webster and Backhouse (eds) 1991, no 47; Tweddle 1992.

9 medieval britain and ireland 2011 known as Mercian Style.40 It can be compared with three similarly shaped, but larger, items from the St Ninian s Isle hoard, though these feature a soldered-on base-plate with cut-outs forming a bar across the centre. This bar might have the same function as those seen on earlier sword-scabbard pyramids.41 Although the Melton example lacks this feature, a small loop on the exterior may have had the same function, perhaps showing a coherent development from the small angular 7th-century pyramids (eg WILT-9AB1A4) through the taller pyramids later in the same century (eg LEIC-4BD061) and those with a polygonal or rounded base (eg IOW-1855E4), culminating in large, conical silver examples. The practical use of such large objects on a scabbard, however, is questionable. Further examples of 8th-century gilded copper-alloy objects include an incomplete cross (Fig 3f), perhaps a brooch, from Somerton (Oxfordshire; BERK-9FA163), and a fragment of a disc brooch or linked pin set from near Cricklade (Wiltshire; WILT-609F56). Two horse-harness mounts of Irish origin were also found at Somerton (Oxfordshire; BERK-E067E3) and Ashburton (Devon; CORN-29D1E2); these may have arrived in England through Viking activity.42 Five more finds of 9th- or 10th-century cubo-octahedral weights were recorded in 2011, including a 3-dot example from Torksey (Lincolnshire; SWYOR-21CA16), which at 2.26 g is the heaviest known 3-dot weight (0.753 g per dot), approximating well to the heaviest weight-per-dot of six-dot and four-dot weights on the PAS database, at g and g per dot respectively.43 Iron objects are rarely recorded by PAS, but two interesting examples were found in An axe from the Longwick area (Buckinghamshire; BUC-B7ACE2) has enough detail surviving to allow it to be dated to the later 10th or 11th century. The blade is asymmetric and the socket has a small projection or spur both above and below, suggesting a Scandinavian origin. Also of note was an early 11th-century iron stirrup from Butlers Marston (Warwickshire; WAW ). Late Anglo-Saxon or Anglo-Scandinavian horse-harness items are relatively common finds, but a fragment of zoomorphic bridle cheekpiece from Newent (Gloucestershire; GLO-49AC57), has extended the known distribution to the west, and an openwork harness pendant in Ringerike style from Skidbrooke (Lincolnshire; PUBLIC-1FCD40) adds to the small number of these pendants recorded on the database. High and late medieval Two hundred and sixty-seven medieval seal matrices were recorded in 2011, including some interesting new finds.44 Several could be tentatively linked to historically attested individuals, including ESS-9419A5 (Henry, son of Henry of Essex; Fig 4e), DENO-4C6163 (John Brocas), BERK-A752A4 (Roger de Moulins) and CORN-622F61 (Thurstan of Treago). A matrix from Cobham (Surrey; SUR-B74173) is the seal of the Augustinian Priory of St Mary and St Wulfade at Stone in Staffordshire; the British Library has a cast of an impression from the same seal.45 Twenty-three papal bullae were recorded during 2011, all dating from c 1200 c 1500 apart from one from the Durham area (DUR-A82CF6), issued by Callixtus II ( ). This is unusual, as all other post-conquest bullae have hitherto all been of Webster 2011, Wilson in Small et al 1973, 122 3; Bruce-Mitford 1959, and n Youngs 2001 gives many parallels to the Irish harness mounts. The preponderance of finds from Somerton is due to the holding of metal-detector rallies in the parish; the findspots are widely spaced. 43 Geake Sixty seal matrices were recorded in Norfolk in 2011, of which only 12 have been recorded on the PAS database. A full catalogue of the Norfolk seals can be found in Rogerson and Ashley Birch 1887, 759, no David Williams (Surrey FLO) has compared the British Library s plaster cast, made c 1850, and SUR-B There is no doubt that the Cobham matrix is the original used.

10 310 portable antiquities scheme Alexander III ( ) or later.46 In addition, a bulla of the Knights Hospitallers was found in Farnham (Essex; BH-8F07F3), issued under Amaury d Amboise, Master of the Order, (Fig 4a). The pilgrims badges found in 2011 were, as usual, more robust than those found by excavation and included several from foreign shrines and cults. Two lead-alloy vernicles, or badges depicting the face of Christ as seen on the Veronica, were found at Swinhope (Lincolnshire; NLM-93AF41) and Drayton Bassett (Staffordshire; WMID- 3585A7). There are now five souvenirs of the Veronica recorded by PAS. All, in theory, would have been purchased in Rome. A copper-alloy badge of St Roch (Fig 4b) found at Somerton (Oxfordshire; SUR- 3AEBC3) is only the third British find associated with the saint,47 who can be recognised by the way his robe is pulled up to expose a sore on his leg, and by the presence of his dog.48 His cult was particularly popular in Britain, but with several major centres on the Continent, the most important being at Venice. A 16th-century silver badge from the Preston area (Lancashire; LANCUM-61F133) shows a female bust wearing a kennelshaped head-dress with the hair loose, and wearing a cross-shaped pendant (Fig 4c). There are striking resemblances to the well-known South Netherlandish wooden bust reliquary of a female saint, thought to be a companion of St Ursula, now in the Metropolitan Museum,49 and the badge may be a souvenir of pilgrimage to Cologne or another site that held relics of St Ursula and her companions. A circular lead-alloy pilgrim badge from Huddleston with Newthorpe (North Yorkshire; SWYOR-25C3E5) appears to be the second pilgrim badge depicting the Fountain of Life known from England. The badge shows Christ crucified between the Virgin Mary and St John, each seemingly holding a vessel up towards him. A similar badge from Salisbury Museum, thought to be a late 15th/early 16th-century Flemish import, shows the blood from Christ s wounds streaming into a circular well or fountain.50 A unique example of a pilgrim s badge with a decorative chequerboard enamel backplate was found at Castle Hedingham (Essex; ESS ). The badge is otherwise unexceptional; its design (showing a half-kneeling saint within a crescent formed from a stag with a cross between its antlers in front of the saint and a horse behind) may represent St Eustace or, more probably, St Hubert.51 Designs of similar shape are also known for St Edmund. It is likely to have been produced in England in the late 15th or early 16th century.52 A curious copper-alloy enamelled plate in the shape of a mounted warrior was found at Stagsden (Bedfordshire; BH-B91E31). The decorative detail includes a coat of arms depicted not only on the shield, but also on the neck and rump of the horse s trapper or caparison (Fig 4d). The blazon has been reconstructed from the tiny traces of enamel, and appears to be: Azure two lions rampant to sinister Argent on a canton Argent in sinister chief a cinquefoil Gules. The mount may relate to the Buckland family, the arms of which were: Gules two lions rampant Argent on a canton Or fret Sable. The lions facing to sinister are unusual, possibly indicating that the arms were accidentally reversed when the mould was made. The crest in the form of a scallop-shell that tops the horse s bridle is matched on a fragment of a similar but larger mount (Fig 4f) from Milden (Suffolk; SF-B15AB4). 46 Three pre-conquest bullae are known; one of Paschal I (817 24) recorded by PAS (HESH-ADE183), and two in Norwich Castle Museum, one of John XI (931 6) from Kilverstone, Norfolk and the other of Theodore I (642 9) from Longwick, Bucks. 47 The other two are both recorded by PAS (SOMDOR905 and YORYM-CEA5E7). 48 Van Beuningen and Koldeweij 1993, Accession number Spencer 1990, 38, no 73, fig Spencer 1990, Robinson 1989, 66 9.

11 medieval britain and ireland fig 4 High- and late-medieval objects. (a) BH-8F07F3 bulla of the Order of the Hospital of St John of Jerusalem. (b) SUR-3AEBC3 pilgrim s badge. (c) LANCUM-61F133 pilgrim s badge. (d) BH-B91E31 mount. (e) ESS-9419A5 seal matrix. (f) SF-B15AB4 mount. Scale 1:1. Photographs: (a) J Watters, (b) D Williams, (c) S Noon, (d) J Watters, (e) L McLean, (f) J Carr. A crushed but near-complete copper-alloy lantern was found near Chippenham (Wiltshire; WILT-837CC4). Missing its top, base and handle, as well as the aperture cover, it is strikingly similar to a probable 14th-century example in the Museum of London.53 A semi-circular copper-alloy sheet with trefoil and quatrefoil perforations found at Wheathampstead (Hertfordshire; BH-40D622) may also be part of a lantern, perhaps part of the top or the aperture cover. 53 Egan 1998, 150 1, fig 118; Ward-Perkins 1940, fig 58.1; accession number A1366.

12 312 portable antiquities scheme An unusual folding candlestick in the Museum of London s collection has now two parallels, from Toft Monks (Norfolk; SF-DB9EE4) and Repps with Bastwick (Norfolk; NMS-6FEA68).54 A figurine from Hartlebury (Worcestershire; WAW-47EC91) may also be part of a candlestick and can be compared to a figurine supporting a pair of candleholders in the Victoria and Albert Museum collections.55 Of less obvious appeal are three lead spindle-whorls which give evidence for manufacturing. Two mould-identical whorls were found within a kilometre of each other on the edge of Leeds (SWYOR-D84976 and SWYOR-D89352) with imitation lettering in relief, while a third, from Arkendale (North Yorkshire; SWYOR-DCDC87), is badly mis-cast. (H Geake)56 research reports A rare Mammen-style rectangular brooch from Lincolnshire The discovery of this unusual rectangular brooch was made by Mr B Wallis in the Market Rasen area (Lincolnshire). The brooch was recorded by Wendy Scott (Leicestershire FLO; LEIC-6AF276) and identified as a type previously recorded by Hattatt, dating to the 10th century.57 It is similar to two other examples from England, from Bergh Apton (Norfolk) and West Stow Heath (Suffolk), bringing the total known for this type to three.58 All three brooches are highly unusual in their combination of Scandinavian and continental cultural influences as well as in their use of Mammen-style decoration, which is rare in Britain. The brooch depicts a semi-naturalistic bird with a composition suggesting movement (Fig 5). The bird is depicted in profile and has a long neck facing downwards, and a head with a round eye. Its diagonally placed wings extend downwards from the back and pass over an upwards-pointing tail; both wings are depicted by double tendrils and are intertwined by a looped tendril emanating from the back of the bird s head. The bird in fact looks like a swan just about to land on the water, with its raised leg and bent foot awkwardly extended in front of it. In the two East Anglian examples, the looped tendril is described as a head lappet as it is depicted firmly attached to the head. These examples also have what is described as an open jaw (unfortunately missing on the Market Rasen brooch). On the reverse are the remains of a single, perforated pin-lug and a broken C-shaped catchplate, both aligned with the brooch rim. The decorative treatment of the bird, particularly the movement and open-ended tendril terminal of the head-lappet and the use of tendril ornament to depict the bird s wings and tail, identify it with the Scandinavian Mammen style.59 It is similar to the compact birds carried on the Bamberg casket and on two horse collars from Søllested, Fyn (Denmark), while a more stylised creature appears on the axehead from Mammen itself.60 Semi-naturalistic birds are also carried over into the next phase of Scandinavian art, the Ringerike style. On stylistic grounds, the brooch can be assigned to the period of currency of the Mammen style: the second half of the 10th century.61 However, whereas the brooch motif is influenced by Scandinavian art-styles, its shape is not Scandinavian in origin. The rectangular brooch shape would appear to relate to rectangular and square brooches found predominantly in northern Germany and the Netherlands: so-called Rechteckfibeln a type also recorded independently in England 54 Ward-Perkins 1940, fig 56.1; accession number A Accession number This object was found in late 2010, and the parallel noted by Geoff Egan just before his untimely death. 56 Department of Archaeology, University of Cambridge, Downing Street, Cambridge CB2 3DZ, England, UK. hg260@cam.ac.uk. 57 Hattatt 1987, 384, fig 125, Ibid; Hinton 1974, 62 3, Pl XVIII, Fuglesang 1991, 84, Ibid, figs 14a,b; 24c and 23b. 61 Fuglesang 1980,

13 medieval britain and ireland fig 5 Mammen-style rectangular brooch (LEIC-6AF276). Scale 1:1 (a), 2:1 (b). Photograph and drawing by W Scott. from, for example, Heighington (Lincolnshire; LIN-5FCA06) and Pakenham (Suffolk; SF-ED1145). Rectangular-shaped brooches are dated on the Continent to the 9th and 10th centuries, with squarer forms, like that from Market Rasen, being late on in the series.62 The Market Rasen brooch thus fits comfortably within this period. Although represented in sculpture in the Isle of Man, the Mammen style is rare on decorative metalwork in Britain and these three brooches provide the only known examples of the style in this medium to date.63 However, the mixing of continental brooch forms with Scandinavian art styles is paralleled by a group of rectangular brooches with Borre/Jellinge-style decoration, all from the Thetford area (Norfolk).64 This unusual blending of cultural styles is of interest given growing evidence from England for the circulation of Carolingian metalwork, including mounts, strap-ends and brooches.65 That such brooches appear to have been manufactured in the Danelaw serves as a reminder of the range of cultural influences at play in this area of 10th-century England. (J Kershaw and W Scott)66 62 Wamers 2000, Abb Mammen-style influences may, however, be present on a Jellinge-style sword chape from York (Fuglesang 1991, cat no 34; Roesdahl et al 1981, YD 41). 64 Eg Wallis 2004, 38, pl V, fig See, for instance, FAHG-123AB4; YORYM-EAF943; DENO ; SF-1E8422; ESS-BE9A25; SWYOR ; NMS-DC Kershaw: Institute of Archaeology, 36 Beaumont Street, Oxford OX1 2PG, England, UK. jane.kershaw@arch. ox.ac.uk. Scott: Finds Liaison Officer Leicestershire and Rutland, Room 400, County Hall, Glenfield, Leicester LE3 8RB, England, UK. wendy.scott@leics.gov.uk.

14 314 portable antiquities scheme Limoges enamels: evangelist symbols on cross terminals Since 1997 the PAS has recorded 172 small copper-alloy mounts, fittings and (less commonly) dress accessories, enamelled in a technique and style known as Limoges,67 after the town where such enamelling was practised from the second half of the 12th century. The technique used is chaplevé enamelling, by which the cells are carved or cast into the surface of a metal object and filled with enamel (made from powdered glass). The object was then fired until the enamel powder melted and flowed to fill the cell. When cold the enamel hardened and the object was polished so that the raised portions of the original surface remained visible, serving to frame the enamelled areas. This sort of enamelling particularly suits copper-alloy objects, which look golden in colour when new. Amongst the Limoges enamels recorded by the PAS are impressive examples of high-medieval art, such as a cross-staff terminal or altarpiece from the Isle of Wight (IOW ) and a pyx lid from near Arncliffe, North Yorkshire (YORYM-58EB27), highlighting the richness of churches prior to the Reformation. Also notable are a strap from a coffret from Albrighton, Shropshire (WMID-247E87), a gilded angel figurine (probably from a reliquary) from Long Newnton, Gloucestershire (GLO-296B71) and a ferrule from Corpusty, Norfolk (NMS-CFA3A4). The remainder consist of various figurines (for caskets, crucifixes and such like), as well as mounts, buckles (some with plates) and terminals for processional crosses, of which the latter form the core of this note.68 Thirteen cross terminals have been recorded by PAS, of which 12 are T-shaped and one trefoil (IOW-541F71; Fig 6a);69 they normally have four rivet holes,70 enabling the mount to be fixed to processional crosses. All the PAS-recorded terminals are decorated with engraved designs for enamelling on the display side (front), though the extent to which the enamel survives varies. They normally show one of the four evangelist symbols, although an example from Knapton, Norfolk (NMS-94BDD2; Fig 6b) depicts a standing figure, possibly St Peter, since he is shown holding a key (of Heaven) in his left hand.71 Following Limoges iconographic tradition the terminals were placed in a predetermined pattern on the reverse arms of crosses (clockwise from top): St John (eagle), St Luke (ox), St Matthew (angel) and St Mark (lion).72 Unfortunately, Limoges enamels recorded by the PAS are often poorly preserved (given they have spent much of their existence probably since the Reformation in the ground) unlike those that survive in (continental) churches or museums, and hence the designs upon them can be uncertain. Furthermore, the distinguishing details of evangelist symbols (such as the horns of the ox 67 Accessed on 24 February Ten of these were found in 2011: BERK-2BA546, BH-83D7E2, BH-BF6A84, GLO-296B71, IOW , NCL-EAF521, SF-B15AB4, SWYOR-167D66, WILT and WMID- 247E For a broader discussion of Limoges enamels recorded by PAS, see Lewis NMS-94BDD2 is distinct from the others since it has a semicircular protrusion at the top to accommodate the saint s halo. 70 There are exceptions, for example NMS-566F60, which is damaged, has at least three remaining fixing holes, and may originally have had more. 71 Shaffery 1998, 14 5, observed that the terminals with evangelist symbols are those most commonly found by metal-detector users, and suggested that Limoges crucifixes in England could have been uniface with the evangelist symbols being transferred to the front of the cross. This terminal of St Peter might therefore be from the obverse of a cross. The devices that decorate cross obverses vary: a cross in the Metropolitan Museum of Art (Boehm and Taburet-Delahaye 1996, 132 3, no 26) has representations of the sun and moon (top), Virgin and St John (left and right) and male saint (bottom), while a cross in Cleveland Museum of Art (ibid, 235) has two angels (top), Virgin and St John (left and right) and male saint (bottom). 72 Shaffery 1999, See also the reverse of the Cross of Bonneval for a complete example (Boehm and Taburet-Delahaye 1996, 315 7, no 104).

15 medieval britain and ireland fig 6 Limoge cross terminals depicting the evangelist symbols. (a) IOW-541F71 St John (eagle). (b) NMS- 94BDD2 St Peter. (c) SUSS-54B2C4 St Luke (winged, nimbate bull). (d) HAMP-A33213 (angel). Scale 1:1. Photographs: (a) F Basford, (b) E Bales, (c) L Burnett, (d) R Webley. of Luke, or beak of the eagle of John) can be unclear due to artistic style,73 though the T-shaped nature of the terminals makes it clear where they were placed on the cross; evangelist symbols are invariably shown the right way up. Only six of the designs on 73 For example IOW-541F71 has been interpreted as St John, though this is not obvious and the terminal is clearly from the bottom of the cross (the place of St Matthew). Likewise, NMS-07E5F4, also interpreted as St John, is on a terminal from the right side of the cross (the place of St Mark); in this case the design is worn and the features that distinguish an eagle from a lion are not as obvious as might be expected. It should be noted that the author has only seen images of these finds, not the actual artefacts which are invariably returned to their finder, and the recorders may have seen details not obvious in a digital image. In the case of NMS-566F60 the recorder has noted that the evangelist symbol is too worn to be identified.

16 316 portable antiquities scheme PAS-recorded cross terminals are certain. Of the top terminals one is St Peter (NMS- 94BDD2), one is St John (NMS-07E5F4) and another, which is worn, is uncertain (NMS-566F60); a further terminal (NMS1177) is recorded as St John but lacks an image and the terminal side is not stated. All three right terminals are of St Luke (Fig 6c; SUSS- 54B2C4, WILT-45FC93 and SF-8AE233). Only one lower terminal is certainly St Matthew (KENT3396), but two others are strong possibilities (Fig 6d; HAMP-A33213 and LON-71F0A3); another, though very worn, is recorded as St John (IOW-541F71). Of the left terminals one is described as St Luke (DOR-C06CB4), which is possible, and another as St John (GLO-B28210), but in both cases the design is unclear. Since most evangelist symbols are winged and their designs are stylistically indistinct (even the diagnostic features may be less obvious than expected), uncertain symbols can easily be designated as eagles (of St John) or angels (St Matthew). PAS finds suggest that placement rules were not always followed, with metalworkers having more licence to place cross terminals than previously thought. A further clue to identifying evangelist symbols (and aiding with the orientation of damaged and fragmentary terminals) is provided by small arrows occasionally found on their reverse (ie non-display side), as found on two PAS-recorded examples. The arrow on the reverse of the St Peter terminal (NMS-94BDD2; Fig 7a) points upwards, and therefore may have aided the cross maker, although presumably the saint would never have been fixed the wrong way up. Arrows were likely added before the display face was designed, as seems to be confirmed by a terminal depicting St Luke (WILT-45FC93; Fig 7b) that has a downward pointing arrow on the reverse: this terminal was originally to be applied to the left-hand side of the cross (in the place of St Mark) but the craftsperson applying the design (of St Luke, the right way up) ignored this, unless all other metalwork pieces on this cross also had downward facing arrows. Unfortunately, the PAS sample is too fig 7 Reverses of cross terminals showing alignment arrows. In both cases the arrows are located inbetween the two top rivet holes. (a) NMS-94BDD2. (b) WILT-45FC93. Scale 1:1. Photographs: (a) E Bales, (b) K Hinds.

17 medieval britain and ireland small to be definitive, but, as future pieces are discovered, so the meaning of such (at first) seemingly insignificant features will no doubt become clearer. (M Lewis)74 A distinctive group of late-medieval keys: a Norfolk speciality? Since 1992 a distinctive group of ten medieval copper-alloy rotary keys has been recorded in Norfolk (Fig 8)75 and an eleventh has been found at Brookland (Kent).76 These share very similar characteristics and are most likely products of the same mould. On each the lozengiform bow is topped by a hollow knop and carries openwork decoration, a cross with a round aperture in each angle and an additional aperture in two corners. The shank is hollow throughout most of its length, collared and of sub-rectangular cross-section, and the bit is complex. Minor surface differences, particularly in the details of openwork apertures, are the result of variations in the post-casting fettling process. One example is decorated with filed grooves and engraved lattice on the knop and collar (Fig 9). Of two keys from London that share a similar hollow knop or tubular loop at the top, one also has a cruciform design in openwork on the bow.77 They both, however, are of Ward- Perkins Type VI, with shanks that are solid except at the end next to the bit. They were dated to the 15th century, but the Norfolk and Kent examples should be more broadly allocated, to the 14th or 15th. This strong emphasis on Norfolk might be skewed by the larger number of metal objects recorded there in the last 30 years than elsewhere. However, the distribution within the county, where recording has been quite evenly applied, does indicate a genuinely localised concentration, with all ten being found in the eastern half and nine in the north-eastern part of the county, with two examples coming from the parish of Hevingham. Standardised, national and to a certain degree international styles of small metal objects were the norm in High Middle Ages, and individual taste is rarely encountered. A rural workshop, or perhaps one in the market town of Aylsham, is most likely the source of these objects. It is hoped that in future further examples of medieval localism will be thrown up by the PAS. (A Rogerson78 and S Ashley)79 Erratum: An Anglo-Saxon burial from West Hanney, Oxfordshire In the above note in Portable Antiquities Scheme 2010 (Medieval Archaeol 55, 294 6), close parallels for the West Hanney brooch were cited as Avent nos from Milton (Kent).80 The location details should have read Milton II, near Abingdon (Oxfordshire). Our apologies for any confusion this may have caused. (J Naylor and A Byard) 74 Department of Portable Antiquities & Treasure, British Museum, London WC1B 3DG, England, UK. mlewis@thebritishmuseum.ac.uk. 75 Aldborough (HER 55452), Bracon Ash (HER 55518), Hainford (HER 50072, PAS NMS-D5ABD1), Hevingham (HER 39947), Hevingham (HER 41687, PAS NMS-9E9D00), Letheringsett with Glandford (HER 33796; PAS NMS-A154D6), Sall (HER 31178), Stokesby with Herringby (HER 41137, PAS NMS-BAD0F6), Swafield (HER 23485) and Woodbastwick (HER 55011, PAS NMS-C92422). 76 PAS KENT Ward-Perkins 1940, 140 1, pl XXIX nos 13 and Historic Environment Service, Union House, Gressenhall, Dereham, Norfolk NR20 4DR, England, UK. andrew.rogerson@norfolk.gov.uk. 79 Address as above. steven.ashley@norfolk.gov.uk. 80 Avent 1975.

18 318 portable antiquities scheme fig 8 Five keys from Norfolk. (a) Bracon Ash. (b) Hevingham. (c) Aldborough. (d) Hevingham. (e) Stokesby with Herringby. Scale 1:1. Photographs: E Whitcombe (a) and (c), E Bales (b) and (e), E Darch (d). Images (a), (c) and (d) Historic Environment Service, Norfolk County Council.

19 medieval britain and ireland fig 9 A key from Hevingham, Norfolk (HER 41687). Scale 1:1. Drawn by J Gibbons. Historic Environment Service, Norfolk County Council. Abdy, R 2005, After patching: imported and recycled coinage in fifth- and sixth-century Britain, in B Cook and G Williams (eds), Coinage and History in the North Sea World. Essays in Honour of Marion Archibald, Leiden: Brill, Allen, M, Leins, I, Moorhead, S, et al (eds) forthcoming, Coin Register 2012, Brit Numis J 82. Avent R 1975, Anglo-Saxon Garnet Inlaid Disc Composite Brooches, Brit Archaeol Rep Brit Ser 11. Birch, W de G 1887, Catalogue of Seals in the Department of Manuscripts in the British Museum, London: British Museum. Boehm, B D and Taburet-Delahaye, E (eds) 1996, Enamels of Limoges , New York: The Metropolitan Museum of Art. Boughton, D, Williams, G and Ager, B 2012, Viking hoards. Buried wealth of the Norse North West, Curr Archaeol 264 (March 2012), Bruce-Mitford, R L S 1959, Comments on the bowls and miscellaneous silver, and general conclusions, in A C O Dell, R B K Stevenson, BIBLIOGRAPHY T N Brown et al, The St Ninian s Isle Silver Hoard, Antiquity 33, , Cook, B 1999, Foreign coins in medieval England, in L Travaini (ed), Local Coins, Foreign Coins: Italy and Europe 11th 15th Centuries, Milan: Società Numismatica Italiana, Collana di Numismatica e Scienze Affini 2, Danet, A and Périn, P 1992, Musée des antiquités, Rouen: de l Egypte ancienne à la Renaissance rouennaise, Rouen: Musée des Antiquités. Egan, G 1998, The Medieval Household: Daily Living c.1150 c.1450, London: The Stationery Office. Egan, G 2007, Byzantium in London? New archaeological evidence for 11th century links between England and the Byzantine World, in M Grünbart, E Kislinger, A Muthesius and D Stathakopoulos (eds), Material Culture and Well-Being in Byzantium ( ), Vienna: Verlag der Österreichischen Akademie der Wissenschaften, Elias, E R D 1984, The Anglo-Gallic Coins, London: Spink.

20 320 portable antiquities scheme Evison, V I 1994, An Anglo-Saxon Cemetery at Great Chesterford, Essex, Counc Brit Archaeol Res Rep 91. Fuglesang, S H 1980, Some Aspects of the Ringerike Style: A Phase of 11th Century Scandinavian Art, Odense: Odense University Press. Fuglesang, S H 1991, The axehead from Mammen and the Mammen Style, in M Iversen (ed), Mammen: Grav, Kunst og Samfund i Vikingetid, Højbjerg: Jysk arkeologisk selskab, Geake, H 2010, Viking-Age cubo-octahedral weights recorded on the PAS database: weights and weight-standards, in J Naylor and H Geake, Portable Antiquities Scheme, Medieval Archaeol 54, , Gilchrist, R 2008, Magic for the dead? The archaeology of magic in later medieval burial rites, Medieval Archaeol 52, Hattatt, R 1987 Brooches of Antiquity, Oxford: Oxbow. Hawkes, S C (with Cameron, E and Hamerow, H) 2000, The Anglo-Saxon cemetery of Bifrons, in the parish of Patrixbourne, East Kent, Anglo-Saxon Stud Archaeol Hist 11, Hendy, M F 1999, Catalogue of the Byzantine Coins in the Dumbarton Oaks Collection and the Whittemore Collection, 4: Alexius I to Michael VIII, , Washington: Dumbarton Oaks Center for Byzantine Studies, Trustees for Harvard University. Hills, C and Penn, K 1981, The Anglo-Saxon Cemetery at Spong Hill, North Elmham, Part II: Catalogue of Cremations, East Anglian Archaeol Rep 11. Hills, C, Penn, K and Rickett, R 1994, The Anglo- Saxon Cemetery at Spong Hill, North Elmham, Part V: Catalogue of Cremations, East Anglian Archaeol Rep 67. Hinton, D A 1974, A Catalogue of the Anglo-Saxon Ornamental Metalwork in the Department of Antiquities, Ashmolean Museum, Oxford: Clarendon Press. Kelleher, R 2010, The English Custom. Bent coins in medieval England, Treasure Hunting Apr 2010, Koch, A 1998, Bügelfibeln der Merowingerzeit im westlichen Frankreich, Mainz: Römisch- Germanisches Zentralmuseum. Lewis, M 2011, Heavenly Treasures: Limoges enamels recorded by the Portable Antiquities Scheme, Treasure Hunting Dec 2011, Menghin, W 1983, Das Schwert im Frühen Mittelalter, Stuttgart: Konrad Theiss Verlag. Milne, G and Richards, J D 1992, Wharram. A Study of Settlement on the Yorkshire Wolds VII: Two Anglo-Saxon Buildings and Associated Finds, York University Archaeol Pubs 9, York: York University. Moorhead, T N S and Naylor, J 2012, Byzantine coins found in Britain, Treasure Hunting Feb 2012, Naismith, R 2005, Islamic coins from early medieval England, Numis Chron 165, Naylor, J 2011a, The circulation of sceattas in western England and Wales, in J Naylor and H Geake, Portable Antiquities Scheme, Medieval Archaeol 55, , Naylor, J 2011b, Focus on coinage in 2010, in J Naylor and H Geake, Portable Antiquities Scheme, Medieval Archaeol 55, , Naylor, J 2012, Early medieval coins found in Britain. Part 1 Merovingians and Visigoths, Treasure Hunting May 2012, North, J J 1994, English Hammered Coinage. Volume 1. Early Anglo-Saxon to Henry III c. 600 to 1272 (3rd rev ed), London: Spink. Robinson, J 1989, A late medieval pilgrim badge from Chaucer House, Tabard Street, SE1, London Archaeol 6. Roesdahl, E, Graham-Campbell, J, Conner, P and Pearson, K (eds) 1981, The Vikings in England and in their Danish Homeland, London: Anglo-Danish Viking Project. Rogerson, A and Ashley S 2012, Medieval seal matrices from Norfolk, 2011, Norfolk Archaeol 46:3. Shaffery, P J 1998, The Limoges Crucifix Project A Further Update, Treasure Hunting Feb 1998, Shaffery, P J 1999, The Limoges Crucifix Project Progress & Information, Treasure Hunting Dec 1999, Small, A, Thomas, C and Wilson, D M, 1973, St Ninian s Isle and its Treasure, London: Oxford University Press. Spencer, B 1990, Salisbury Museum Medieval Catalogue Part 2: Pilgrim Souvenirs and Secular Badges, Salisbury: Salisbury and South Wiltshire Museum. Spink 2003, Coins of Scotland, Ireland and the Islands (Jersey, Guernsey, Man and Lundy). Pre-decimal Issues (2nd ed), London: Spink. Tweddle, D 1992, The Anglian Helmet from Coppergate, Archaeol York 17:8, York: CBA/York Archaeological Trust. Van Beuningen, H J E and Koldeweij, A M 1993, Heilig en Profaan; 1000 laat-middeleeuwse insignes, Cothen: Stichting Middeleeuwse Religieuze en Profane Insignes.

21 Wallis, H 2004, Excavations at Mill Lane, Thetford, 1995, East Anglian Archaeol Rep 108. Wamers, E 2000 Karolingerzeit, in R Müller (ed), Reallexikon der Germanischen Altertumskunde. Fibel und Fibeltracht, Berlin: Walter de Gruyter, Ward-Perkins, J B 1940, London Museum Medieval Catalogue, London: London Museum. Weber, M 2000, Das sächsische Gräberfeld von Issendorf, Landkreis Stade, Niedersachsen, Teil 2 = Studien zur Sachsenforschung 9:2. Webster, L 2011, Style: influences, chronology, and meaning, in H Hamerow, D Hinton and S Crawford (eds), The Oxford Handbook of medieval britain and ireland Anglo-Saxon Archaeology, Oxford: Oxford University Press, Webster, L and Backhouse, J (eds) 1991, The Making of England: Anglo-Saxon Art and Culture AD , London: British Museum Press. Williams, G 2001, Coin-brooches of Edward the Confessor and William I, Brit Numis J 71, Youngs, S 2001, From Ireland Coming : fine Irish metalwork from the Medway, Kent, England, in C Hourihane (ed), From Ireland Coming: Irish Art from the Early Christian to the Late Gothic Period and its European Context, Princeton: Princeton University Press, MEDIEVAL BRITAIN AND IRELAND FIELDWORK HIGHLIGHTS IN 2011 editor s note: This section comprises a selection of highlighted projects or discoveries made across Britain and Ireland in 2011, with extended illustrated overviews of results, finds and contexts. The selection each year gives scope to identify results and findings from (sometimes ongoing) fieldwork (whether excavation, landscape or buildings archaeology) which has regional or national significance in medieval archaeology. Contributors are welcome to offer potential extended summaries (of maximum 2,000 words with short supporting bibliography where appropriate; good quality b&w and/or colour plates and figures, to a maximum of four). The deadline for submissions is 1 March each year. MB&I entries (with illustrations) of traditional format are still required; since 2008, these are housed and accessible on an online database and directory hosted by ADS and accessible via the Society s webpages. Thus the Index listed below uses the numbers to signify the full sequential county and country listings; however, only those numbers in bold refer to entries that appear in print in this volume as Highlights. The searchable database will eventually include all back issues and entries of MB&I. We thank all contributors past, present and future for their efforts, texts and support in maintaining this invaluable source of summaries on annual fieldwork activities and discoveries. The Editor extends his gratitude to Märit Gaimster, who is stepping down as MB&I British compiler after many years of excellent service, and likewise to Kieran O Conor and Rory Sherlock for their full and conscientious work on the Irish entries. For future submissions, please contact: British sites Tiziana Vitali, Pre-Construct Archaeology Ltd, 9 Cranfield Road, Brockley Cross, London SE14 1TN, England, UK; tvitali@pre-construct.com Irish sites Tomás Ó Carragáin, with Patrick Gleeson, Department of Archaeology, University College Cork, Ireland; T.OCarragain@ucc.ie; papgleeson@gmail.com

22 322 fieldwork highlights INDEX FOR MEDIEVAL BRITAIN AND IRELAND 2011* I. PRE-CONQUEST agriculture: 194. amber artefacts: 264. assembly site/thing site: 267, boats: 282. bone/antler artefacts: 74, 103, 298. boundaries, ditches, enclosures: 118, 144, 200, 213, 255, 264. buildings: 178, 264, 298. burials, cemeteries: 99, 118, 195, 255, 282, 303. ceramics: 9, 20, 29, 31 2, 36, 38, 44, 47, 65, 85, 95, 99, 101 2, 112, 115, 140, 147, 155, 167, 175, 180 2, 186, 194, 213, 231, 264, 270, 284. churches and chapels: 23, 268. coins: 99. earthworks: 11. environmental evidence: 178. glass (vessel, window): 264, 298. glass bead: 310. hacksilver: 290. industrial sites: ceramic: 231. jet: 39. metal: 155, 160, 181 2, 260, 262, 264. textile: 39, 102, 155. leather: 74. metal artefacts: 20, 155, 160, 255, 264, 278, 282, 283, 290, 292, 295, 298. monastic sites: 103, 321, 331. ovens, kilns: 255. Pictish features: 264, 297 8, 301, 303, 311, 317, 321. ringfort, hillfort: 256, scientific dating (archaeomagnetic, dendro, radiocarbon): settlements: 6, 140, 160, 255, 264, 278, 282, 283, 290, 292, 295, 298. settlements, urban: 74, 85, 102. souterrain: 256, 284. stone artefacts: 231, 282. stone carving: 264. stone, funerary (cross-slabs, grave-markers, gravecovers, head supports): 301, 317. sunken-featured buildings: 9 10, 39, 132, 155, tool set: 282. Viking-period features: 103, 267, 278, 282, 285 6, 290, waterworks: 147. weapons: 282 II. POST-CONQUEST agriculture: 194. almshouse: 334. barns: 57. boats, ships: 86, 208. bone/antler artefacts: 74, 79, 81, 244, 279. boundaries, ditches, enclosures: 8, 21, 90, 141, 144, 183 4, 200, 205, 215, 218, 254. buildings, civic/public: 274. buildings, defensive (peel towers, tower houses): 53, 272, 296. buildings, domestic: 62, 102, 128, 136, 142, 150, 158, 189, 197, 213, 319. burials, cemeteries: 28, 34, 63, 76, 113, 176, 222, 234, 247, 251, 328. castles: 149, 164, 175, 186, 214, 244, 246, 250, 273, 287, 289, 291, 333. cathedrals: 252. ceramic building material: 29, 42, 61, 67, 71, 145, 175, 275, 330. ceramic floor tile: 3, 39, 63, 83, 113, 258, 330. ceramics (pottery, tiles); 1, 4 5, 8, 14, 19 20, 22, 24 6, 33, 35 40, 44, 46, 53 5, 58, 60, 64 5, 67, 71 72, 78, 80 1, 85, 88, 90 1, 93, 95, 99, , 112, , , 123, 125, 140 1, 144 8, 152, 156, 159, 162, 165, 170 1, 173 5, 187 9, 192 4, 196 7, 200, 204, 209, , 224 5, 227 8, 232, 237, 241, 244, 248, 253 4, 257 8, 263, 270 1, 277, 279, 310, , 330, 332, 334, 337. cesspit, latrine, garderobe: 10, 176, 185, 217. churches and chapels: 23, 28, 50, 63, 66, 82 4, 102, 111, 113, 151, 193, 212, 243, 245, 251, 259, 268, 294, 308, 320, 336, 338. coins, jetons: 310, 313, 318, 322 3, 328, 334. coin weight: 314. college, school: 50, 221 2, 334. deer park: 89. deserted medieval village (DMV): 197, 242. dovecote: 14, 235, 242. earthworks: 11, 45, 137, 242. environmental evidence: 157. farmsteads: 68, 124. field systems: 7, 13, 15 16, 24, 26 27, 30, 45, 48, 55, 60, 69 70, 109, 122, , 134 5, 138 9, 153 4, 156, 163, 166, 168 9, 177, 180, 184, 191 2, 198 9, 201 2, 206, 240, 242, 300. fish weir: 281. gaming board: 329. gardens: 333. * Nos in bold indicate their inclusion as Highlights following this Index.

23 glass (vessel, window): 63, 99, 244. grange: 51. hearths, fireplaces: 142. hospitals: 97, 222. industrial sites: baking: 125. butchery: 58. ceramic (pottery, tiles): 12, 64, 171 2, 204, 275. lime: 133, 246. metal: 6, 219, 249, 253, 277. mortar: 249. salt: 60. shipbuilding: 86. unspecified: 41. kitchen: 211. leather: 73 75, 78, 86, 172, 249. manors and moated sites: 6, 14, 31, 143, 156, 165, 230, 242, 265. metal artefacts: 6, 20, 59, 81, 86, 99, 175, 244, 253, 283, 302, 304 6, , 321, 324 7, , 332, 335. mills (wind): 25. monastic sites: 3, 34, 49, 56, 61, 83, 92, 95, , 111, 119, 170, 179, 212, 253, 258, 263, 269, 280, 293, 307, , 321, 324 7, , 332, 335. Norman features: 252. ovens, kilns: 39, 117, 133, 157, 324. palaces, ecclesiastical: 296. palaces, royal: 104. quarrying: 10, 40, 93, , 127. reliquary: 212. medieval britain and ireland refuse, domestic (pits, middens): 60, 67, 78, 175, 196, 244, 248 9, 254, 277. roads, holloways and streets: 52, 126, 146, 238, 242. Saxo-Norman features (including ceramics): 75, 78 9, 99, 145, 175, 179, 182. scientific dating (archaeomagnetic, dendro, radiocarbon): 246, seal matrix: 306. settlements: 2, 4 6, 10, 12, 17 18, 37, 71, 115, 122, 125, 140, 152, 185, 187, 189, 220, 226 7, 239, 254, 263, 271, 299. settlements, urban: 41 43, 72 87, 96 7, 102, 110, 176, 190, 196, , 221 4, 228 9, 233, 236 7, 248 9, 277, 279, 334. shrine: 329. smithy: 219, 253. stone, architectural: 79, 84, 186, 243, 266, 269, 280, 293, , 326, 329. stone, funerary: 329. sundial: 3. surgical instruments: 244. textiles: 86. town defences: undercroft: 176. villages: 137, 171, 174, 180, 207, 276. voussoir: 269. waterfronts: 86. waterworks (conduits, dams, drains, ponds, tanks): 147, 244. weights, scales: 99, 253. wells: 129. woodland, forests: 288. yards, metalled surfaces: 10, 94. ENGLAND NORTH YORKSHIRE QUESTIONING MEDIEVAL SETTLEMENT DESERTION IN THE WOLDS Mitchell Pollington (Archaeological Services WYAS) 242. hanging grimston, near kirby underdale (SE ). In 2011, English Heritage commissioned Archaeological Services WYAS to carry out a detailed earthwork survey and investigation of the deserted medieval settlement of Hanging Grimston.1 This was undertaken to assess the form and condition of the surface remains, which had not previously been recorded in detail, and to inform the future management of the site. It was followed in early 2012 by an extensive geophysical survey2 funded as part of the LEADER Coast, Wolds, Wetlands and Waterways (CWWW) programme, financed by the European Agriculture Fund and the Department for Environment, Food and Rural Affairs (Defra). The results of the surveys at Hanging Grimston have provided the first steps in attempting to understand the site s form and development. Further archaeological investigations, including documentary research, additional geophysical survey and targeted 1 Pollington ASWYAS forthcoming.

24 324 fieldwork highlights excavation, will be undertaken on the site in the near future as part of a Heritage Lottery Fund funded community-based project organised by the High Wolds Heritage Group. Located on the NW edge of the Yorkshire Wolds, c 20 km north-east of York, Hanging Grimston comprises an extensive and complex series of earthworks, representing elements of medieval and post-medieval settlement. The site is situated on a steep southfacing slope and consists of a number of fields of permanent sheep pasture. The survey area encompasses c 22.6 ha around Mount Pleasant Farm, including the whole of the Scheduled Monument area (SM no 32665; NHLE no ) and adjacent fields, which also contain earthwork remains. Brief history of the site The earliest documentary reference to Hanging Grimston is from Domesday Book and, as with many of the villages in this part of Yorkshire, it is recorded as being waste at the time,3 probably as a result of the Harrying of the North by William I. In the late 1080s the manor was granted to St Mary s Abbey, York,4 and there are various documentary references relating to it from the early 13th century onwards.5 The Poll Tax returns of 1381 provide the earliest details of the occupants of Hanging Grimston and record a taxable population of 79.6 A process of enclosure and depopulation appears to have begun by at least the early 16th century, when an Inquisition of 1517 records the enclosure of 40 ha, with eight people driven from the village and two houses devastated,7 presumably to make way for sheep pasture. Later documents, especially a number of letters recording a dispute between the Abbot of St Mary s and Thomas Cromwell over the amount of pasture available in the mid-1530s,8 highlight the growing importance of sheep to the economy in the area. Following the Dissolution, Hanging Grimston was granted to Edward Fiennes, Lord Clinton,9 and by the late 16th century it was held by the Bourchier family. Despite the earlier enclosure, a Feet of Fines of 1575 records that a substantial settlement still existed at this time, apparently comprising forty cottages and a watermill, although these may have been spread throughout the township.10 By 1619 a Chancery Petition shows that there was only a limited area of arable remaining, and it is likely that any settlement on the site had been reduced to a small number of farms.11 The planned medieval village (Figs 1 2) The remains of a planned village are visible to the south and south-west of Mount Pleasant Farm, consisting of two rows of tofts surviving as earthwork platforms, aligned N S down the hill on both sides of a substantial holloway. The holloway itself survives up to 21 m wide and 2.5 m deep at its S end, but is disturbed by later activity along its centre before continuing northwards as a narrower hollow which joins the line of the modern road to the north. The most northerly row of tofts is situated on the W side of the holloway and consists of at least five broadly rectangular platforms, each measuring c 25 m wide and 42 m long, and sharing a common frontage. A 2.5 m wide back lane runs along the rear of the tofts and is separated from the fields to the west by a headland. There is no surface evidence for croft boundaries extending to the rear of these tofts, although any remains of such boundaries may have been obscured by later medieval ploughing. Two building platforms 3 Farrer Leadam 1893, Eg Ellis 1883, 11 and PRO 1891, Beresford 1952, Leadam 1893, Shepherd 1928, National Archives ref LR 15/ Collins Beresford 1952, 62.

25 medieval britain and ireland fig 1 Hanging Grimston: oblique aerial photograph looking north-east. Photograph English Heritage NMR. are cut into the E end of the southernmost tofts, measuring only around 4 m wide and 6 m long; these appear to be too small to contain houses. The sites of two larger buildings, each about m long and up to 10 m wide, are aligned across two of the tofts to the north and are represented by level platforms with wall lines also identified as parch marks. Any continuation of this row to the south has been obscured by the presence of a later probable farmstead. The remains of a S row of tofts are situated outside the Scheduled Monument area, aligned along the E side of the holloway. This row comprises seven platforms, although these vary in size and plan between m long and m wide, and earthwork evidence for any buildings is slight. The platforms do not share a common frontage, although a back lane is aligned along their rear. There is again no evidence for associated crofts, although ridge-and-furrow to the east appears to over-plough the back lane, and this ploughing could have destroyed any surface evidence for earlier boundaries. It seems possible that a further row of tofts could have been situated on the E side of the holloway, opposite the N row, although there is no earthwork evidence for this. However, this area, to the immediate south of Mount Pleasant Farm, has been heavily disturbed by later activity, including the presence of a late-medieval or post-medieval farmstead and more recent drainage work, which could easily have destroyed such remains. To the east of this, a series of E-W aligned boundaries define a number of long plots, which were separated from the open fields by a ditch or narrow lane. These plots appear similar in form to crofts found to the rear of toft platforms in other deserted settlements on the Yorkshire Wolds and could be associated with the missing tofts to the west.

26 326 fieldwork highlights fig 2 Hanging Grimston: plan of the earthwork survey. Drawing Archaeological Services WYAS.

27 medieval britain and ireland The manorial enclosure The N side of the survey area is occupied by a substantial sub-rectangular enclosure, the size and scale of which suggest that this represents a manorial site. The enclosure is formed by a boundary consisting of an alignment of prominent banks and scarps, up to 2.5 m high along its S side. These define a sub-rectangular area measuring at least 110 m N S and 140 m E W, covering approximately 1.6 ha. A low section of bank aligned N S outside the enclosure s W side lies on a slightly different alignment to the enclosure boundary and may represent a different phase of development. The interior of the enclosure is divided into two main areas by internal E-W boundaries, as well as into smaller plots or compounds within these. These contain evidence for the remains of a number of structures, including the site of a rectangular building close to the enclosure s S edge. This is aligned N S and measures 14 6 m, the walls of which survive as low banks. On the N side of the enclosure are the well-defined remains of two rectangular buildings in an L-shaped configuration; the walls of both buildings are defined by banks surviving up to 0.3 m high and 2.5 m wide. Just to the west is a large circular bank, surrounding a central depression, c 11.5 m in diameter, which could be the site of a dovecote. Later farmsteads The well-preserved earthwork remains of a farmstead lie in the field to the south of Mount Pleasant Farm, east of the line of the holloway, and these clearly overlie boundaries relating to the earlier medieval village. The farmstead consists of a large U-shaped building range, probably representing a farmhouse, measuring m, with a yard to the south. A large rectangular platform to its north may denote an associated barn. Another probable farmstead is situated on the W side of the holloway, to the north-east of View Cottage, although the remains are less clearly defined. A level area to the south may be an associated yard, on the NW side of which are the earthwork remains of a building. An extant stone-built post-medieval barn, now associated with View Cottage to the immediate east, has a blocked doorway in its E side suggesting that it may have originally required access to this area, and perhaps itself formed part of the earlier farmstead. Phasing the earthworks It is tempting to see the rows of tofts at Hanging Grimston originating as part of a reorganisation of the village in the late 11th or early 12th centuries, following the upheavals of the 1060s and the granting of the manor to St Mary s Abbey in the late 1080s. However, they could equally relate to later re-planning, and the rows are not necessarily contemporary in origin. The farmsteads that overlay the earlier village remains are similar to the type of courtyard farms that appear in the Wolds by the 16th century, suggesting that the form of village had changed substantially by this time. Other stratigraphic relationships between surface features are unclear, often due to later disturbance. For example, a modern farm track along a low embankment has severed any visible relationships between the N row of tofts and the possible manorial enclosure to the north. However, the fact that the E side of the enclosure curves around to follow the holloway suggests that this route was well established and still in use when the enclosure was laid out. Geophysical surveys (Fig 3) Both magnetometer and resistivity surveys were undertaken on the site; these largely confirm the plan of the medieval village recorded as earthworks and parch marks, while also identifying the likely position of buildings within the S row which could not be recognised on the ground. The magnetometer survey also recorded previously unexpected evidence for a series of enclosures beneath the extant ridge-and-furrow on the W side of the site, which appear to comprise a possible Iron-Age or early Roman period ladder settlement focused on a routeway curving up the hillside. Similar enclosures have been identified

28 328 fieldwork highlights fig 3 Hanging Grimston: plan of the magnetometer survey. Image Archaeological Services WYAS.

29 medieval britain and ireland underlying the earthworks to the east of the existing holloway, and one can speculate that this route has its origins in the later prehistoric period. Indeed, both the holloway and the routeway, identified as part of the geophysical survey, appear to have influenced the alignment of the extant curving field boundaries on the N side of the survey area, where these two routes meet and continue northwards along a deep holloway along which runs the modern road. The geophysical surveys have therefore opened up the possibility that settlement at Hanging Grimston has late prehistoric roots, and that this earlier landscape had a direct influence not just on the plan of the medieval settlement but even the post-medieval field pattern. FINE DINING AND WATER SUPPLY AT A LATE-MEDIEVAL PLEASAUNCE IN WENSLEYDALE Erik Matthews (The Architectural and Archaeological Society of Durham and Northumberland) 244. hornby castle (SE ). Hornby Castle lies approximately six miles to the north-west of Bedale in Wensleydale. The site is partially ruinous following demolition of the N and E wings and the Great Hall after the loss of the estate by the 11th Duke of Leeds in 1930 (Fig 4). On the basis of a reference in Leland s Itinerary the Castle is conventionally assumed to have been built by William Ist Baron Conyers in the early years of the 16th century. Since summer 2010 the Architectural and Archaeological Society of Durham and Northumberland have been engaged in a programme of fieldwork involving field walking, building recording and excavation at the site, following on from earlier fig 4 Hornby Castle: plan showing the trench locations relative to the main Castle building (pre-1930 demolition), based on the First Edition 6-inch OS Map.

30 330 fieldwork highlights building recording at the surviving house in 2006/07. This earlier work provided evidence for a scheme of heraldic decoration as well as surviving elements of the castle chapel, which challenge the conventional interpretation of a development of the site in the late 15th/early 16th centuries, associated with William Ist Baron Conyers. The evidence instead suggests a significant period of expansion of the site during the period of ownership of Sir John Conyers KG, a major political and military figure on the national stage in the mid-15th century. Season 1 of the excavation focused on a 5 7 m trench adjacent to the ruined late 18th-century Banqueting House, located some 450 m to the south-west of the main castle building. Along with a considerable quantity of material associated with the mid-18thcentury and subsequent refurbishments of the castle, quantities of late-medieval worked stone and part of an early architectural terracotta frieze were recovered. At the end of the season some traces of a detached late-medieval formal garden, including part of a stonerevetted flowerbed, a sand path, a canalised water course and the mortar floor of an open-sided building were uncovered beneath the later material in the base of Trench 1. This posed a number of important questions in respect of the development of the site, which Season 2 has sought to answer. The 2011 excavations Work in Season 2 focused on a second 6 8 m trench (subsequently extended) set 35 m north-west of Trench 1, and closer to the boundary of the site with the church of St Mary. The location was chosen since surrounding trees and vegetation were showing evidence of die-back and, during fieldwalking in the previous year, quantities of high-status medieval pottery were recovered. Indeed, at an early stage of Season 2, a young volunteer recovered a 15th-century whalebone dice of Flemish manufacture close to the W edge of the trench. Close to the surface of Trench 2 a NE SW gravel path associated with the 18thcentury designed landscape was recorded. This yielded a single sherd of an 18th-century embossed glass jar from its surface, although further quantities of residual late-medieval (mainly 14th-century) pottery was recovered from the remainder of the trench to the north-west. The path had a well-made foundation course which significantly included a piece of late-medieval clay roof tile, subsequently identified as originating in Flanders. Directly adjacent to the path, towards the S edge of the trench, three large pieces of worked stone were identified which had clearly been robbed from a substantial stone wall in the vicinity. Subsequently, at the NE edge of the trench, a 1.5 m wide section of in-situ stone surface was uncovered. Associated with this partially robbed wall, crossing the trench in a NE SW direction, were significant quantities of food waste in the form of animal bone, many exceedingly large. The bone was mainly from deer and boar, but other more exotic species such as crane, peacock, beaver and deep sea and coarse fish were also present. A few centimetres inside the line of the wall at the NE edge of the trench were found a copper-alloy stylus and the pin from a silver-alloy penannular brooch, marked with the initial letter E and the symbol of a sunburst. In view of its scale it was initially thought that the wall was a boundary wall, or at the very least an outer wall of a building; however, subsequent discoveries indicate it to be an internal spine wall. Removal of a dead pine tree close to the NE edge of the trench enabled an extension to the trench, exposing two floor surfaces of medieval date, one timber and one of mortar. From here, two dress buckles dating from the late 14th/early 15th centuries were recovered together with a parchment pricker in apple wood. A further timber floor overlain with mortar was recorded at the N edge of the trench; a razor with a bone handle was recovered from its surface. Over the remainder of the trench, either side of the wall, was an uneven, coarse mortar surface through which a considerable number of timber stake holes had been inserted. These have been interpreted as a means to support a suspended timber floor an element known from elite residences of the late 14th/15th centuries. Further animal

31 medieval britain and ireland bone was recovered together with small fragments of imported glass ware, a very large quantity of fragments of meat-serving ceramic vessels, some of which were imported, together with upwards of 60 handles from ceramic wine jugs made in the Humber Ware kilns and the Gainsborough area of Lincolnshire, with some imports from the Saintonge region of S France. More unusual finds include a barrel padlock, a rowel spur of mid-15th-century date and a surgical instrument for bleeding known as a fleam. The water system Probably the most unexpected discovery at Hornby so far has been the water-supply system recorded running W E across the trench and then parallel with the stone wall in a NE SW direction. This took the form of a pressurised piped water supply in elm pipes together with a formalised drainage system laid in ceramic pipes; subsequent to its construction a link was formed between the two to allow for the drain to flush out. The water system was first identified in the form of a substantial hollowed-out tree trunk with a much narrower pipe leading off to the south-east. A small leak caused by the excavator nicking the side of the pipe with his trowel proved the water still to be running through it under pressure. The drainage system laid in ceramic pipes was first discovered directly to the east of the central spine wall (Fig 5). It had been laid in a mortar bedding to minimise leakage, although conclusive evidence that this was indeed a drainage system did not come until later. A ceramic pipe lying roughly central within the trench was discovered to link the piped water system to the ceramic pipe work, and beyond the join towards the S end of the trench water was identified as flowing through the ceramic pipe. A further significant length of ceramic pipe lay at the N end of the trench and beyond and included a join with another smaller section of ceramic pipe towards the south-east which had been fig 5 Hornby Castle: ceramic drain and mortar floor from the north-east. Photograph Erik Matthews.

32 332 fieldwork highlights broken and then rather crudely blocked with part of an 18th-century field drain. While the lower levels of the trench extension were being excavated, a pungent almost sulphurous smell was noted, and when the new section of pipe was uncovered it became clear that this was due to leakage from the pipe through the extremely crudely made joints; at certain locations a significant build-up of green cess was observed which included small bones of fish and poultry. There was even evidence of a series of timber-lined inspection chambers at various points within the system. Importantly, sherds of early to mid-14thcentury pottery came from the construction cut for the ceramic pipe system. Hornby Castle as a late-medieval residence The new discoveries at Hornby of an earlier phase of occupation than previously recorded raise the question of who the owner was of the late-medieval building, and what role it played as a residence. The answer may be sought in the relative wealth and national political connections of several of the owners of Hornby at this time. The finds assemblage recovered so far indicates a use centred on entertainment, primarily feasting but with evidence of music and gaming taking place as well. The origin of some of the material is of importance, with sherds of Bohemian glassware and red-clay roof tiles originating in Flanders among the items recovered. The various floor surfaces with associated pottery indicate two key phases of occupation: one in the early 14th century and the other in the mid-15th century, both associated with owners of significant wealth and courtly connections. In the early 14th century the castle belonged to Sir John Neville, a household knight of King Edward III, and later his brother Sir Robert Neville who was part of the household of Edward the Black Prince. Sir John was closely involved with the so-called Nottingham Coup of 1330 in which King Edward seized power from his mother and her lover the Earl of March, for which he was amply rewarded. We may tentatively associate construction of the work at Hornby with that time. Sir Robert was also closely involved in the early stages of the Hundred Years War before being a casualty at the siege of Calais in Of potentially more significance is the usage of the site in the mid-15th century when Hornby was owned by Sir John Conyers KG, a man who played a pivotal role in the regional and national political scene during the reign of King Edward IV.12 Indeed, there is documentary evidence of a visit by the King to the site in December It is to this period that much of the excavated pottery and more especially the artefacts from Season 2 can be ascribed. Sir John held high office in the extended household of the Nevilles of Middleham and also the Royal administration over a sustained period of time; furthermore, he is recorded on several occasions in the 1470s engaging in diplomatic missions to Scotland and also to Burgundy. The structure uncovered at Hornby Castle, particularly bearing in mind the technical sophistication of the water-supply and drainage arrangements, may be tentatively identified as a pleasaunce or place for the entertainment of elite guests well away from the hub of the principal residence, comparable to examples at Kenilworth, Warwicks, and Bodiam, Sussex.13 The possession of such a facility would have been of great importance to anyone with the wealth and aspirations to hold and retain high office and political influence in such a volatile period as the mid-15th century, and can be seen very much as being at the business end of power and influence.14 Season 3 (2012) will aim to uncover more of the building and to seek further understanding of its relationship with both the adjacent deserted medieval village and the main castle complex. Already the possibility exists that the castle may have formed part of a deliberately created skyline of towers, in the fashion suggested by Phillip Dixon for Wingfield in Derbyshire at this time Dockray 1986, Pamela Marshall pers comm. 14 Goodall 2011, Dixon 2011,

33 medieval britain and ireland fig 6 Hornby Castle: Andenne Ware jar rim sherd, early 12th century. Photograph Erik Matthews. Postscript Preliminary analysis of the pottery in 2012 has identified a significant residual deposit of sherds of earlier 12th-century Andenne Ware (Fig 6), a high-status fineware originating from NE France. While occurring in some quantity in S and SE England, it is extremely rare in the region of the excavation. Might this deposit relate to an earlier building associated with elite entertainment? This hypothesis will be explored in future work at Hornby Castle. SCOTLAND HIGHLAND A VIKING AT REST: NEW DISCOVERIES ON ARDNAMURCHAN Oliver Harris (University of Leicester), Hannah Cobb (University of Manchester), Héléna Gray (CFA-Archaeology Ltd) and Phil Richardson (Archaeology Scotland) 282. ardnamurchan, swordle bay (NM ). This entry recounts the exciting discovery of an intact Viking boat burial in Ardnamurchan, western Scotland. The burial is the first of its kind excavated by archaeologists on the UK mainland and so forms an important contribution both to our understanding of the Viking relationship with Scotland and of the broader regional connections with the Viking diaspora.16 The Ardnamurchan Transitions Project (ATP)17 is a long-term research- and community-led initiative investigating the occupation of the Ardnamurchan peninsula, the most westerly part of 16 Graham-Campbell and Batey <

34 334 fieldwork highlights the British mainland. Although the area around Ardnamurchan, particularly the Inner Hebrides, has long been the subject of extensive archaeological research on all periods, Ardnamurchan itself has seen little work. The ATP, which began in 2006, is designed to address this and to contribute to local and regional, as well as national and international understandings of a series of key transitions. Fundamentally, ATP seeks to examine the long-term occupation of the peninsula from the Mesolithic through to the present day, through an analysis of Swordle Bay on the N coast in the first instance, using field survey, geophysical analysis, environmental sampling and excavation. The Project is thus interested in the multi-scalar nature of historical process and the relationship between these different scales, and, accordingly, our excavations have focused on sites from different periods. These include the Neolithic chambered cairn of Cladh Aindreis, which has demonstrated evidence for Bronze-Age reuse; the nearby Bronze-Age kerbed cairn named Ricky s Cairn (previously unknown before the Project s work); the Iron-Age promontory fort of Dun Mhurchaidh; and the 19th-century settlement of Swordle Corrach. Each site has provided evidence for numerous different kinds of social practice, as well as local and regional connections, and together these provide a narrative for long-term change within Swordle Bay. The Viking boat burial Part of the field survey work has included recording all possible above-ground archaeology within the bay. It was in this survey work that we initially identified the site that we describe in this short report. On a lower terrace of the Swordle River, a small area of ridge-and-furrow field system was observed to respect a discrete circular turfed stone mound, which appeared to be artificially raised above the natural lie of the land on its W (river-facing) side. In 2006 a trench measuring 1 5 m (with a 2 1 m extension to the NW end) was opened to investigate this structure and examine its relationship with the surrounding ridge-and-furrow. Excavation suggested that this was a small circular structure, set into the beach gravels, but one subject to more recent robbing activity. At the time there was little to suggest the richness of the archaeology that was to follow. The excavation was duly recorded in our data structure report for that year,18 and it was planned to return to this site in future seasons. Between 2007 and 2010 the project concentrated on its excavations elsewhere in the Bay, returning to the site in It was this season that revealed the extraordinary deposits we report here. Initially the L-shaped trench described above was re-opened and all backfilled deposits removed. The trench was then extended to a total size of 6 3 m to the east; this covered much of the mound that had been initially identified. Topsoil overlay a large spread of medium to large, angular to sub-angular stones. Some of these appeared to be in situ and were termed cairn stones, while others appeared to have been dragged away from the central feature and had clear marks on them from being struck by a plough. Given the evidence for later lazy bedding surrounding the feature, many stones defining this had clearly been dislodged by ploughing. Exposing the stones in this way indicated that we did not have the full extent of the feature, and so the trench was extended m to the south, thereby forming a T-shape. Once recorded, the stones were removed to reveal an oval-shaped feature, measuring m at its widest point, and oriented WSW ENE. It was defined by a series of kerb stones set within a cut into the small mound of beach shingle, and filled with other stones (Fig 7). The clear boat-shaped nature of the stone-filled feature immediately raised the possibility this was a Viking boat burial. Indeed it paralleled Aith on Fetlar on Shetland, excavated by TimeTeam in 2002, and the famous boat burial of Balladoole on the Isle of Man.19 The stones contained within the cut had not been placed at random; 18 Richardson and Cobb Aith: Colleen Batey, pers comm, and Val Turner in The Shetland Times, 30 August 2002; Balladoole: Bersu and Wilson 1966, 1 44.

35 medieval britain and ireland fig 7 Ardnamurchan: pre-excavation shot of the boat burial. Photograph Ardnamurchan Transitions Project. instead there were two clear outer layers of deliberate kerbing. The stones in the centre appeared to have been less ordered, if with some attempt to define a central burial area. Clearly there had been significant disturbance of the site after burial, not least in the rotting of the boat and other organic elements, and so the location of many of these stones would no doubt have shifted. As these central stones were recorded and removed the first finds began to appear, including a spear (Peterson type E)20 and a shield boss. A thin layer around the boss was identified as the decayed remains of the shield itself. Interestingly further stones underlay these artefacts suggesting they may have been inserted during the filling of the burial, as a succession of deposits. The spear was evidently broken in antiquity perhaps marking a ritual killing of this weapon (see also the sword below). Close to the base of the feature an increasing number of finds was recovered, alongside numerous in-situ rivets belonging to the boat itself (these and all finds were threedimensionally recorded). Finds included a whetstone of Norwegian schist,21 a complete sickle and a copper-alloy ringed pin. The latter was initially identified as Irish in origin but recent re-examination by Dr Colleen Batey suggests that there is only a single equivalent example in Fanning s corpus of Irish examples,22 making the origin of this pin potentially Scandinavian. Further weapons were recovered close to the base of the feature, comprising an axe (broad-bladed in form) and a sword (Fig 8). This latter find has proven to be of particular interest: it was slightly bent, perhaps deliberately before deposition, and thus has similarities with the treatment of the spear noted above; it may have been wrapped in textiles prior to deposition (although these could also have adhered to the sword as a 20 Colleen Batey, pers comm. 21 Alsvik with Batey Fanning 1994.

36 336 fieldwork highlights fig 8 Ardnamurchan: the sword in context. Photograph Ardnamurchan Transitions Project. result of close contact with the Viking s clothing or shroud), which have survived through the mineralisation of the sword itself. The high-quality pommel has silver- and copper-wire decoration forming a checkered pattern, and the grip appears to be bone/horn (Fig 9). Conservation by Pieta Greaves of AOC Archaeology in Edinburgh is ongoing, and further information is anticipated. In terms of typology, Dr Gareth Williams of the British Museum has suggested that it may be a transitional form, a variant of a Petersen Type K, which tends to have five straight pommel lobes and straight hand guards, and a later Type O, with flared pommel lobes but also curved or straight guards.23 This could suggest a date in the late 9th to early 10th centuries ad an argument supported by Dr Batey. Further finds include an unusual ladle with a handle over one metre in length. This is paralleled only by an example from the Kiloran Bay burial on Colonsay. Within this ladle had been placed a hammer and a set of tongs, plus organic materials that have yet to be fully identified (a similar set of tools, though without the ladle, but with the addition of an adze, has been recovered from Ballinaby, Islay).24 Close by was the copper-alloy top of a drinking horn. Finally, a set of potential strike-a-light flints came from the very base of the boat. After full excavation the shape of the cut could be properly appreciated. Clearly boat-shaped, not just in plan but in section, it measured m with steep sloping, slightly concave sides (Fig 10); the prow and stern ends of the cut were more gradually sloping. The base was narrow and flat. Through the three-dimensional recording of over 200 rivets, along with the series of profiles taken through the cut, the full shape of the boat 23 Cf Peirce 2002; Petersen Graham-Campbell and Batey 1998, 123.

37 medieval britain and ireland fig 9 Ardnamurchan: close-up of the decoration on the sword pommel. Photograph Ardnamurchan Transitions Project. can hopefully be reconstructed. The mineralised wood preserved around many of the rivets (and next to other finds) may also add detail in terms of species used and about manufacturing techniques, beyond the fact the boat itself was certainly clinker-built. Finally, what of the Viking whose burial this was? Although the conditions preserved much of the metalwork and led to the mineralisation of wood from the boat, unfortunately the same cannot be said for the deceased, of whom only two teeth and two small scraps of bone survived. However, isotopic analysis from the teeth may yet give us a glimpse into where the individual spent the early years of his life, while wear analysis may guide on his diet. Wider significance The discovery of an intact Viking boat burial is obviously of great significance, both to the ATP and to wider studies of the Viking diaspora, and particularly their relations with the British mainland. In many ways Ardnamurchan is not an unexpected place to find this kind of burial. Technically it is connected to the mainland, but in many ways it is island-like, and has close connections through time with Mull to the south and Eigg, Rhum and Muck to the north. It is thus not surprising that the burial and its finds link with the well known Viking burials in the Inner Hebrides, and with examples from Orkney, such as Scar, than with other burials on the mainland.25 The closest parallel is undoubtedly the boat burial at Kiloran Bay on Colonsay which is comparable in terms of its richness and in terms of the unusual ladle recovered.26 Nevertheless, on the UK mainland, 25 Cf Owen and Dalland Graham-Campbell and Batey 1998.

38 338 fieldwork highlights fig 10 Ardnamurchan: post-excavation shot of the boat burial. Photograph Ardnamurchan Transitions Project. potentially the only other boat burial is from Huna in Caithness, where a scatter of rivets was recorded in the first part of the 20th century. For our Project, the site also fills in an important hole in our chronology. Analysing long-term change through time is a lot easier if you have something to fill the 1,900-year gap between our Iron-Age promontory fort and our post-medieval settlement! The discovery has significance in various other ways. The important organic and artefactual preservation, coupled with the meticulous excavation of the site under modern conditions, will enable us to develop a detailed understanding of the burial that is matched only by the other modern examples from Scar and Westness in Orkney.27 In turn, this will allow us to make a broader series of interpretations regarding the funeral rite itself. The likely ritual killing of key artefacts in the assemblage (reflected in the possible deliberate bending of the sword and the definite breaking of the spear), and the potential wrapping of the sword in its own shroud (indicated by the mineralised textile on the scabbard of the sword), already hint at some of the performances that accompanied the act of burial. Moreover, the find of the ladle and the as yet unidentified organic remains within it may provide substance to suggestions that such burials included acts such as the pouring of libations.28 Developing a better understanding of the performances that may have surrounded this Viking boat burial will ultimately offer significant insights into not just the buried man but also the beliefs and acts of those living communities who put him to his final rest. There are clearly many facets still to be resolved about the burial. Conservation is ongoing and extensive post-excavation analysis will follow this, including the identification 27 Graham-Campbell and Batey 1998; Owen and Dalland Graham-Campbell and Batey 1998.